#gemrb@irc.freenode.net logs for 3 Oct 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:55:20] <Edheldil> good night
[02:08:29] --> Darki has joined #gemrb
[02:08:30] --- ChanServ gives channel operator status to Darki
[02:09:02] <-- Dark-Star has left IRC (Read error: 60 (Operation timed out))
[03:10:53] <-- exultbot has left IRC (signing off...)
[03:12:18] --> exultbot has joined #gemrb
[03:12:18] --- Topic for #gemrb is: GemRB 0.5.1 | http://gemrb.sf.net | Be wary of your words for there are Modron sensors in this channel: http://log.usecode.org/gemrblog.php | Hey <CHARNAME>, we need some awesome screenshots!
[03:12:18] --- Topic for #gemrb set by lynxlynxlynx at Thu Aug 27 14:07:31 2009
[03:22:02] <-- barra__out has left IRC ("Verlassend")
[04:29:56] <-- wjp has left IRC (kornbluth.freenode.net irc.freenode.net)
[04:32:19] --> wjp has joined #gemrb
[06:59:57] --- ChanServ gives channel operator status to wjp
[07:50:32] --> lynxlynxlynx has joined #gemrb
[07:50:32] --- ChanServ gives channel operator status to lynxlynxlynx
[09:24:21] <CIA-22> gemrb: 03lynxlupodian * r7328 10/gemrb/trunk/gemrb/plugins/Core/ (Actions.cpp Actor.cpp Map.cpp): improved the Hide action with Taimon's findings
[09:35:44] <fuzzie> morning
[09:37:20] <lynxlynxlynx> oj
[09:39:03] <fuzzie> very nice of Taimon to provide python for the light check :)
[09:39:51] <Edheldil> ohajo godzaimas
[09:42:46] <fuzzie> so we have two showstoppers left, Forge_On which Avenger will maybe do, and the final iwd cutscene failing?
[09:43:21] <lynxlynxlynx> sure, but the second one is not as important
[09:48:14] <fuzzie> i am just looking for things to do :)
[09:49:59] <fuzzie> any thoughts? maybe the visibility for scripting
[09:50:22] --> Gekz has joined #GemRB
[09:52:55] <-- tombhadAC has left IRC (Read error: 104 (Connection reset by peer))
[09:54:27] <lynxlynxlynx> not a bad choice at all
[09:55:30] --> tasos86 has joined #gemrb
[09:59:04] <Gekz> hi
[09:59:13] <Gekz> http://superiorchicken.info/1005ha-linux.png
[10:03:18] <CIA-22> gemrb: 03avenger_teambg * r7329 10/chitem/trunk/ (9 files): support gemprjtl.ids in dltcep
[10:03:47] <CIA-22> gemrb: 03avenger_teambg * r7330 10/gemrb/trunk/gemrb/override/pst/gemprjtl.ids: PST:fixed some projectile ids
[10:04:34] <lynxlynxlynx> bought a new one Gekz?
[10:04:54] <Gekz> yeah
[10:04:55] <Gekz> 1005HA
[10:05:01] <Gekz> I had a hackintosh on it for a while
[10:05:09] <lynxlynxlynx> or are you just pointing out the sloppy interpretation of sophomoric? ;)
[10:05:10] <Gekz> but then I joined the pirate party and I'm the current spokesman so...
[10:05:15] <Gekz> breaking EULAs is not on the agenda anymore
[10:05:15] <Gekz> lolo
[10:05:29] <Gekz> lynxlynxlynx, a spaniard likely did that
[10:05:35] <Gekz> look at the gnome bar at the top
[10:05:38] <Gekz> and you'll see
[10:05:43] <Gekz> there's no menubar in any app
[10:05:44] <Gekz> it's mac-like
[10:05:55] <Gekz> http://superiorchicken.info/1005ha-mac2.png
[10:07:56] <fuzzie> so how does gemrb run?
[10:08:02] <Gekz> on this?
[10:08:04] <Gekz> havent tried
[10:08:07] <Gekz> I'd imagine pretty well
[10:08:22] <Gekz> I wonder when Syllable OS will get OSS emulation
[10:08:30] <Gekz> because then we can have a "GemRB Appliance" so to speak
[10:08:39] <Gekz> :)
[10:11:10] <fuzzie> from what they said on your thread, i think they might just fix openalsoft to work directly against Syllable?
[10:12:53] <Gekz> no
[10:12:54] <Gekz> OSS
[10:13:24] <fuzzie> boring :(
[10:13:29] <Gekz> lol
[10:13:30] <Gekz> why
[10:14:06] <lynxlynxlynx> oss4?
[10:23:11] --> xrogaan has joined #gemrb
[10:24:12] <Gekz> no idea
[10:24:16] <lynxlynxlynx> i see why the actor centering doesn't work
[10:24:24] <lynxlynxlynx> you were right fuzzy :)
[10:24:39] <fuzzie> can you fix it? :) it's incredibly annoying
[10:25:15] <lynxlynxlynx> GameControl::ChangeMap does schedule a change and then disables the SF_CENTER flag, but then the tilemap ends up refreshing it with other coordinates
[10:25:20] <lynxlynxlynx> i'm trying
[10:29:48] <-- Gekz has left IRC ("Leaving")
[10:30:04] --> Gekz has joined #GemRB
[10:30:58] <-- Gekz has left IRC (Client Quit)
[10:31:15] --> Gekz has joined #GemRB
[10:31:54] <-- Gekz has left IRC (Client Quit)
[10:32:14] --> Gekz has joined #GemRB
[10:34:19] <-- tasos86 has left IRC ("Leaving")
[10:42:30] <-- Gekz has left IRC ("Leaving")
[10:42:44] --> Gekz has joined #GemRB
[10:56:58] <lynxlynxlynx> done :D
[10:57:59] <Gekz> what's done
[10:58:15] <lynxlynxlynx> fixed the centering
[10:58:19] <lynxlynxlynx> just testing it more now
[11:03:36] <CIA-22> gemrb: 03lynxlupodian * r7331 10/gemrb/trunk/gemrb/plugins/Core/TileOverlay.cpp:
[11:03:36] <CIA-22> gemrb: when drawing, only bump the viewport if it isn't already on the move;
[11:03:36] <CIA-22> gemrb: fixes the viewport not centering on the party when entering new areas
[11:05:53] --> |Cable| has joined #gemrb
[11:06:44] <fuzzie> nice
[11:06:55] <fuzzie> everything really improves :)
[11:21:49] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[11:23:44] --> Gekz has joined #GemRB
[11:26:39] <-- xrogaan has left IRC ("Why ?")
[11:27:01] <fuzzie> hmm
[11:27:13] <fuzzie> trying to install some mods is not so easy
[11:27:20] <fuzzie> i want to install dungeon-be-gone, how do i do it?
[11:27:29] <lynxlynxlynx> it's not a weidu mod?
[11:27:32] <fuzzie> the 'dungeon-be-gone' download gives an .exe, the 'dungeon-be-gone OS X' download gives a .dmg
[11:27:44] <fuzzie> both are weidu inside, but that's not particularly helpful
[11:27:50] <lynxlynxlynx> the exe is usually just a selfextracting zip
[11:28:02] <fuzzie> this one isn't :/
[11:28:14] <fuzzie> maybe it's just not weidu on windows
[11:28:36] <fuzzie> no, it is, it ends up with a .tp2. grr.
[11:28:56] <fuzzie> maybe i should start a campaign for people to put damn zip files up for their mods :)
[11:29:08] <fuzzie> for now let me copy this from someone else's install
[11:29:22] <lynxlynxlynx> some do actually :)
[11:30:08] <fuzzie> yes, but far from all :(
[11:30:13] <fuzzie> the g3 mods are usually v.good about it, at least
[11:30:23] <fuzzie> it turns out that dungeon-be-gone comes with an .exe inside the weidu data, heh
[11:30:39] <fuzzie> (for the ogg decoder)
[11:33:50] <lynxlynxlynx> meh, another barbarian bug
[11:34:03] <lynxlynxlynx> turns out their backstab immunity really is hardcoded
[11:34:57] <Gekz> fuzzie: use 7zip to open it?
[11:35:05] <Gekz> lynxlynxlynx: haha that's horrible
[11:35:37] <fuzzie> Gekz: hey, that worked :D
[11:35:42] <fuzzie> oh, no, it didn't
[11:35:51] <Gekz> aww
[11:36:00] <Gekz> fuzzie: why not
[11:36:02] <fuzzie> just the windows machine finally got around to copying the files in, heh :p
[11:36:04] <Gekz> link me to it
[11:36:09] <fuzzie> Gekz: it's some custom mod .exe installer thingie
[11:36:13] <Gekz> link me to it
[11:36:14] <Gekz> lol
[11:36:16] <fuzzie> http://mods.pocketplane.net/DungeonBeGoneV16.zip
[11:36:19] <Gekz> I will get the shit out of it
[11:36:25] <Gekz> it is my challenge now
[11:36:47] <lynxlynxlynx> :)
[11:37:08] <fuzzie> i have just copied the 'Setup-DungeonBeGone.tp2' and 'jassy' files out of someone else's bg2 install on the network
[11:37:17] <Gekz> sadface
[11:38:26] <Gekz> fuzzie: 7zip worked for me
[11:38:34] <fuzzie> Gekz: but then you get an .exe out, no?
[11:38:37] <Gekz> no
[11:38:43] <Gekz> I unzipped the .exe file
[11:38:44] <Gekz> lol
[11:38:48] <Gekz> i got a jassy dir
[11:38:50] <fuzzie> huh. doesn't work for me. :(
[11:38:57] <fuzzie> the jassy dir is the right thing, though
[11:39:01] <Gekz> and
[11:39:15] <Gekz> Setup-DungeonBeGone.exe
[11:39:21] <Gekz> etup-DungeonBeGone.tp2
[11:39:25] <Gekz> dbgaudio.txt
[11:39:32] <Gekz> and jassy was in $_OUTDIR
[11:41:15] <fuzzie> i wonder why it doesn't like me
[11:41:35] <Gekz> you suck?
[11:41:35] <Gekz> lol
[11:41:37] <Gekz> 7z x
[11:41:51] <Gekz> 7z x DungeonBeGone-v16.exe
[11:41:55] <fuzzie> ye, that's what i did
[11:41:57] <fuzzie> maybe my 7zip is old?
[11:41:59] <Gekz> what version
[11:42:01] <Gekz> I have 9.04
[11:42:18] <fuzzie> hm, it says 4.65
[11:42:23] <lynxlynxlynx> heh
[11:42:30] <Gekz> you're old and insignificant
[11:42:33] <Gekz> *nods*
[11:42:34] <fuzzie> apparently
[11:42:59] <fuzzie> hm
[11:43:08] <fuzzie> ok, gemrb segfaulted :(
[11:44:15] <fuzzie> nasty camera brokenness, also
[11:44:45] <fuzzie> don't think from lynx's changes, i didn't rebuild yet
[11:45:18] <lynxlynxlynx> they only postpone movement anyway
[11:45:37] <fuzzie> in fact, i didn't even svn update yet
[11:46:18] <fuzzie> lynxlynxlynx: this looks like something is postponing camera movement too much
[11:46:38] <fuzzie> but i changed some of that code recently, so probably my fault
[11:47:04] <lynxlynxlynx> my changes only prevent movement orders overwritting each other
[11:47:17] <lynxlynxlynx> but as you said, you don't even have that code yet
[11:48:46] <fuzzie> hm
[11:49:00] <fuzzie> aren't bg2 chars meant to have fists or something if you have no weapon?
[11:49:10] <lynxlynxlynx> yes
[11:49:25] <fuzzie> but we don't support that in the action bar?
[11:49:48] <lynxlynxlynx> it's a bit buggy
[11:51:18] <fuzzie> hm
[11:51:44] <fuzzie> so under gemrb, dungeon-be-gone shows up some giveitem bug, i guess?
[11:51:55] <fuzzie> gemrb fails to give feedback when dropping stuff on the floor
[11:52:03] <fuzzie> and the mod gives the items to the wrong characters anyway
[11:52:59] <fuzzie> however, it does manage to jump me through the dungeon, hoorah
[11:53:37] <fuzzie> and the day/night change stuff completely breaks, heh
[11:53:47] <fuzzie> i was also the last to change that, i think
[11:55:49] <fuzzie> heh
[11:56:01] <fuzzie> ok, so 'import' in the chargen gets the wrong portrait, i guess
[11:56:22] <fuzzie> my Record screen in bg2 now has a huge portrait covering most of the gui
[11:56:40] <lynxlynxlynx> we give feedback for full inventory drops
[11:56:58] <fuzzie> well, gemrb dropped a bunch of stuff on the floor and gave me no feedback
[11:57:07] <lynxlynxlynx> maybe i missed an action
[11:57:26] <lynxlynxlynx> or maybe the mod does it that way itself
[11:57:33] <Gekz> fuzzie: so this mod tests a lot of features of the engine, does it?
[11:57:37] <fuzzie> lynxlynxlynx: GiveItemCreate
[11:58:11] <fuzzie> hm
[11:58:20] <fuzzie> actually, GivePartyAllEquipment is probably where it fails
[11:58:38] <fuzzie> picks the wrong actor and all
[11:59:09] <fuzzie> it needs to go forward through the party, and it needs to give feedback
[11:59:18] <fuzzie> think you can fix, lynx, if you added the feedback in the first place?
[12:00:25] <fuzzie> same problem exists for MoveInventory and Plunder and TakeItemList and so forth, i guess
[12:01:18] <fuzzie> strange though, MoveItemCore looks like it should do the feedback
[12:01:37] <fuzzie> maybe this is a more subtle bug, hum
[12:03:05] <lynxlynxlynx> appears so, i just used the target already available in the functions
[12:03:47] <fuzzie> we don't do the voice yet, i guess?
[12:04:37] <fuzzie> right, yes, sorry, the feedback is a dialog bug
[12:05:29] <lynxlynxlynx> just missing verbal constants, I guess
[12:05:44] <fuzzie> oh?
[12:06:07] <lynxlynxlynx> you got the text, but no voice complaints?
[12:06:32] <fuzzie> the code doesn't pass an actor to the string display function
[12:06:57] <lynxlynxlynx> i'm not sure if this sound is not the same for everyone though
[12:07:04] <fuzzie> it's per-actor
[12:07:49] <fuzzie> i mean, i don't know how this works, sorry, but Minsc and Yoshimo and Imoen and etc have different "my inventory is full!" sounds, is what I mean
[12:08:15] <lynxlynxlynx> yeah, so it's a verbal constant thing
[12:08:27] <fuzzie> and we don't support those yet?
[12:08:39] <lynxlynxlynx> the "dropped" message likely has no sound attached
[12:09:28] <lynxlynxlynx> we're just missing something like: DisplayStringCore(this, VB_INVFULLDROP, DS_CONSOLE|DS_CONST );
[12:11:02] <CIA-22> gemrb: 03fuzzie * r7332 10/gemrb/trunk/gemrb/plugins/Core/Actions.cpp: GivePartyAllEquipment: don't go through party backwards
[12:11:17] <fuzzie> ok, so no easy fix.
[12:11:28] <lynxlynxlynx> the strings there usually contain comments, eg. a critical hit would display "Critical Hit!", while the vb would have "Take that!" and a sound
[12:11:39] <lynxlynxlynx> no, it should be easy
[12:15:07] <fuzzie> so, what's gemrb's vorbis support useful for?
[12:15:58] <lynxlynxlynx> my guess would be mods not needing to convert them to wav
[12:16:05] <fuzzie> well, gemrb only looks for the .wav file
[12:16:11] <fuzzie> so :(
[12:16:27] <lynxlynxlynx> i guess it is unfinished
[12:17:22] <fuzzie> i guess OpenALAudioDriver only tries with IE_WAV_CLASS_ID
[12:18:06] <fuzzie> and there's no equivalent for .ogg
[12:20:35] <fuzzie> any idea who wrote this? i'd like to just add an _OGG_CLASS_ID to try, but i don't know who i'd ask
[12:21:13] <lynxlynxlynx> wjp perhaps
[12:22:09] <wjp> hm?
[12:22:57] <fuzzie> well, i was wondering about the whole system
[12:23:01] <fuzzie> but it seems balrog wrote that
[12:23:07] <fuzzie> wjp added the bit i was interested in, which is the png override
[12:23:29] <fuzzie> i guess i really do just want IE_OGG_CLASS_ID
[12:24:37] <wjp> it's currently hidden in ACMImp it seems
[12:25:19] <fuzzie> wjp: yes, but the stream already got opened at that point
[12:25:27] <fuzzie> so it doesn't find the .ogg file
[12:26:04] <wjp> right
[12:26:18] <wjp> I don't particularly like the whole input plugin system
[12:26:24] <wjp> it doesn't make much sense
[12:27:10] <wjp> the theory of allowing different filetypes for certain types of resources makes sense, but it doesn't seem to be implemented at the right levels
[12:27:14] <fuzzie> yes :/
[12:27:26] <fuzzie> the theory to say "i want a sound file, name <x>" is good
[12:27:34] <fuzzie> but the reality is you have to say "i want a sound file from plugin <y>"
[12:28:36] <wjp> at some point I was planning to put wrappers in GameData
[12:28:40] <fuzzie> and now i look at includes/SClassID.h to try adding an IE_OGG_CLASS_ID and, help?
[12:29:42] <wjp> and there's something in KeyImp::GetFactoryResource that does some png/bmp trickery
[12:29:48] <fuzzie> yes, you wrote that :-)
[12:30:03] <fuzzie> or svn blames you to the beginning of time for it, at least
[12:30:04] <wjp> yes, very ad-hoc
[12:30:16] <fuzzie> i figured i'd just add a similar hack to the OpenAL driver
[12:30:17] <wjp> yeah, I'm fairly sure I added that png thing
[12:30:36] <fuzzie> but i don't know what on earth the numbers in SClassID mean
[12:31:16] <fuzzie> i shall pick a random one
[12:31:32] <wjp> there seem to be three sequences of numbers
[12:31:48] <wjp> kind of
[12:31:51] <fuzzie> namely 0x7
[12:32:11] <wjp> I know of no reasons not to
[12:34:05] <fuzzie> this will be a hack since i won't provide a plugin for this, but it works beautifully anyway
[12:34:46] <fuzzie> or, well, it doesn't, but it reads and decodes the .ogg and i assume the static coming out of my speakers is an endianism issue :)
[12:40:30] <fuzzie> yep, works now
[12:42:04] <CIA-22> gemrb: 03fuzzie * r7333 10/gemrb/trunk/gemrb/plugins/ACMImporter/readers.cpp: fix big-endian vorbis decoding
[12:43:39] <CIA-22> gemrb: 03fuzzie * r7334 10/gemrb/trunk/gemrb/ (3 files in 3 dirs): add IE_OGG_CLASS_ID hack to check for .ogg files after .wav ones
[12:44:02] <fuzzie> so you can just copy audio/*.ogg from DBG and it'll work, magic.
[12:44:20] <lynxlynxlynx> :)
[12:53:43] <-- Gekz has left IRC (Read error: 110 (Connection timed out))
[12:54:47] --> Gekz has joined #GemRB
[13:08:51] --> barra_library has joined #gemrb
[13:12:42] <CIA-22> gemrb: 03lynxlupodian * r7335 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: fixed core's barbarian detection and added their hardcoded backstab immunity
[13:17:25] <Gekz> meowmeow
[13:57:00] --> Avenger has joined #gemrb
[13:57:04] --- ChanServ gives channel operator status to Avenger
[13:57:08] <Avenger> hi
[13:58:40] <lynxlynxlynx> oj
[14:11:33] <fuzzie> hi, Avenger
[14:28:54] <CIA-22> gemrb: 03fuzzie * r7336 10/gemrb/trunk/gemrb/plugins/Core/Interface.cpp: detect the correct EndianSwitch value at runtime
[14:29:04] <fuzzie> ^- someone should test that
[14:33:31] <Avenger> how?
[14:33:42] <fuzzie> build it, run gemrb, see if it still works :)
[14:34:14] <Avenger> ok
[14:35:00] <fuzzie> Have ye no ears to hear?! It is as Tiax said! DESTINY!
[14:37:50] <Avenger> tiax bit you? :)
[14:38:08] <fuzzie> Tiax is great :)
[14:38:24] <Avenger> he is the greatest
[14:38:34] <Avenger> ok, so, gemrb runs
[14:38:45] <fuzzie> ok, good news :) thankyou
[14:38:51] <Avenger> famous last word: now i flip the EndianSwitch flag
[14:39:34] <fuzzie> i don't override the config setting, of course :)
[14:39:43] <Avenger> i see an endless stream of invalid seeks
[14:39:53] <Avenger> how is it different from before
[14:40:09] <fuzzie> now you should never have to change the flag in the config
[14:40:20] <Gekz> sexy
[14:40:22] <Gekz> how'd you do that
[14:40:27] <Avenger> ahh, you mean, the endianswitch flag just forces it?
[14:40:31] <fuzzie> yes
[14:40:34] <Avenger> ok
[14:40:57] <Avenger> you should still keep it, for very odd setups
[14:41:03] <fuzzie> i just found a few old forum posts from people who forgot to change the flag
[14:41:08] <fuzzie> so i thought i might as well make it default to the right thing
[14:41:22] <Avenger> yep, it is a great feature
[14:41:33] <fuzzie> Avenger: maybe you have an idea why the large portrait doesn't work right when i 'import' in bg2 chargen?
[14:41:40] <Avenger> another would be to detect case sensitivity of the filesystem
[14:41:57] <Avenger> you can do that too :)
[14:42:09] <fuzzie> i get a GUIREC screen with a huuge portrait covering everything :)
[14:42:35] <Avenger> because some characters got a screwed up portrait
[14:42:56] <Avenger> if they work in the original game, then maybe the script needs to do some magic
[14:43:09] <Avenger> GetPortrait, check last character, replace if needed
[14:43:12] <fuzzie> well, i try with 'fighter', it works fine with the ToB fighter ('tobfight' i think)
[14:43:18] <Avenger> yes
[14:43:22] <fuzzie> ok :) i will look later
[14:43:30] <Gekz> well
[14:43:39] <Gekz> I'd like to port ext4 to be case-tolerant
[14:43:41] <Avenger> i noticed that too, but i didn't try to see them in the original game
[14:43:46] <Gekz> as in, it stores the case
[14:43:51] <fuzzie> hehe, i am still amused by Dungeon-Be-Gone now i made the ogg files work, it is great
[14:43:55] <Gekz> but it allows accessing SomeThing as SOMETHING or something
[14:44:19] <lynxlynxlynx> Gekz: fuse
[14:44:20] <fuzzie> Gekz: the trouble is that things break if you trys :)
[14:44:27] <lynxlynxlynx> why limit yourself to one choice
[14:44:44] <Gekz> lynxlynxlynx: hey?
[14:44:49] <Gekz> fuse allows this?
[14:44:59] <lynxlynxlynx> sure
[14:45:05] <fuzzie> fuse is slow :(
[14:45:10] <Gekz> I'm not sure I understand fuse
[14:45:12] <Gekz> other than its slow
[14:45:12] <Gekz> lol
[14:45:16] <Gekz> especially on mac
[14:45:29] <CIA-22> gemrb: 03lynxlupodian * r7337 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: added a hack so swashbucklers can't backstab
[14:45:34] <lynxlynxlynx> maybe on mac, i don't have speed problems with it
[14:45:49] <fuzzie> lynxlynxlynx: you're not poking at the iwd end cutscene at all?
[14:45:56] <lynxlynxlynx> nope
[14:46:08] <Avenger> ?
[14:46:08] <Avenger> how ogg and dbg connect?
[14:46:08] <Avenger> i thought it is just skipping the initial dungeon
[14:46:15] <Gekz> lynxlynxlynx: youve obviously never compared the same fs in kernel to fuse
[14:46:29] <fuzzie> Avenger: it comes with an NPC who teleports you, and the NPC's voice is in .ogg format
[14:46:59] <Avenger> oh, i see, a lot of mods use ogg as a transfer format
[14:47:02] <fuzzie> yes :)
[14:47:06] <Gekz> being all open and what not
[14:47:18] <fuzzie> they run oggenc.exe to convert to wav, but that doesn't work so great in linux, so i just copy the ogg files instead
[14:47:24] <Avenger> they recode them to wavc, losing a lot of quality
[14:47:32] <Avenger> yep, that's a good stuff
[14:47:37] <CIA-22> gemrb: 03lynxlupodian * r7338 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp:
[14:47:37] <CIA-22> gemrb: check the base stat for backstabibility, so nobody bad gets this ability when
[14:47:37] <CIA-22> gemrb: using an item that gives a bonus
[14:47:39] <lynxlynxlynx> Gekz: so?
[14:47:42] <Avenger> ogg is smaller and the recoding quality loss is gone
[14:47:54] <lynxlynxlynx> me and lets say 1000 users of sorcery are happy with it
[14:48:09] <Gekz> lynxlynxlynx: which filesystem in fuse are you talking about
[14:48:19] <lynxlynxlynx> it's almost just an interface, what you do in the layer is the important bit
[14:48:26] <lynxlynxlynx> castfs
[14:48:31] <Gekz> neverheard of it
[14:48:32] <lynxlynxlynx> unionfs
[14:48:38] <Gekz> I know aufs and unionfs
[14:48:42] <Gekz> but those arent truely filesystems
[14:48:44] <Gekz> they're layers
[14:48:46] <Avenger> castfs is surely some sorcery stuff :)
[14:48:47] <Gekz> that point to kernel modules
[14:49:14] <lynxlynxlynx> of course
[14:49:24] <fuzzie> fuse works a lot better if you're just implementing a layer, yes
[14:49:29] <Gekz> ntfs-3g
[14:49:30] <fuzzie> and i guess that's what you suggested for ext4 :)
[14:49:34] <Gekz> which implements the whole fs
[14:49:42] <Gekz> is slow as fuck
[14:49:47] <Gekz> fuzzie: OH
[14:49:48] <lynxlynxlynx> for case sensitivity, you don't need much more
[14:49:49] <Gekz> makes sense
[14:49:50] <Gekz> haha
[14:49:54] <Gekz> I dont know why that didnt click
[14:49:55] <Gekz> -.-
[14:50:06] <Gekz> that could likely be used for any filesystem then
[14:50:44] <Avenger> lynx why do you need to check for the base stat, if you already check the thief level
[14:51:47] <lynxlynxlynx> items
[14:52:22] <lynxlynxlynx> if an item has the IE_BACKSTABDAMAGEMULTIPLIER increasing effect and is usable by someone that is not a thief, he'd be able to backstab
[14:52:30] <lynxlynxlynx> even swashies would
[14:53:07] <Avenger> would they still see the wrong multiplier on screen?
[14:53:26] <Avenger> i think it would be safer to not increase the multiplier if there is no base
[14:53:31] <lynxlynxlynx> Gekz: http://wiki.winehq.org/CaseInsensitiveFilenames
[14:53:44] <fuzzie> well, if you can't backstab, it doesn't matter what the multiplier shows
[14:53:53] <fuzzie> but this would be simple to check in the original
[14:55:24] <Avenger> wow, hmm, the backstab modifier opcode seem to be fishy in the IE
[14:55:38] <Avenger> it seems it cannot have a duration
[14:55:44] <Avenger> or 'while equipped'
[14:55:50] <Gekz> lynxlynxlynx: I know of CIOPFS
[14:55:58] <Gekz> but it's really half-assed
[14:56:07] <Avenger> it can be only permanent (1 or 9) timing
[14:56:17] <Avenger> if you use with anything other than 1, it converts itself to 9
[14:56:42] <lynxlynxlynx> huh
[14:57:58] <fuzzie> hm, if this is maybe buggy in bg2 then please bear the other games in mind :)
[14:59:02] <Avenger> uh oh, and there is a percentage option too, but it seems it modifies the same field. like a=a*p1/100; (but since it is permanent, it will be reapplied)
[14:59:10] <Avenger> i gotta check this :)
[14:59:12] <-- Avenger has left IRC ("bye!")
[15:02:29] <fuzzie> lynxlynxlynx: the "nobody starts to talk" spellhold bug should be fixed also?
[15:13:30] <lynxlynxlynx> i didn't try
[15:17:18] --- Darki is now known as Dark-Star
[15:17:24] <lynxlynxlynx> yep
[15:20:09] --> Avenger has joined #gemrb
[15:20:13] --- ChanServ gives channel operator status to Avenger
[15:21:49] <Avenger> i was wrong, it replaces 1 with 9, doesn't change the others
[15:22:39] <lynxlynxlynx> :)
[15:40:50] <Avenger> meh, crash in GenerateAction
[15:42:43] <Avenger> actionstable->FindString crashes
[15:42:52] <Avenger> i guess something didn't work with the merging :P
[15:42:54] <fuzzie> no
[15:43:02] <fuzzie> the merging comment says, you may not use GenerateAction
[15:43:07] <fuzzie> that's why i said you have to use a bcs
[15:43:23] <Avenger> but but, i use MoveToPoint
[15:43:32] <Avenger> i mean, my actors want to move :)
[15:43:48] <fuzzie> oh, it crashes randomly?
[15:44:06] <Avenger> well, probably random, yeah
[15:44:17] <fuzzie> i mean, i'm sure i didn't break walking :p
[15:44:20] <fuzzie> i played a lot since then
[15:44:29] <Avenger> i dont' recall if i tried to walk after the change
[15:45:04] <Avenger> this is the normal movetopointformation call when i click around
[15:45:32] <Avenger> i don't know why is it crashing, because all variables are nonzero, and seemingly intact
[15:45:55] <Avenger> it looks like the function pointer table itself is corrupt or such
[15:45:59] <Avenger> some memory overwriting
[15:46:16] <fuzzie> hm, are you on Windows?
[15:46:19] <Avenger> yes
[15:46:26] <Avenger> no valgrind here :)
[15:46:36] <fuzzie> maybe the ogg change broke something
[15:46:52] <Avenger> this one-->int i = actionsTable->FindString(String, len);
[15:47:06] <Avenger> heh, i doubt it was the ogg
[15:47:12] <Avenger> it is more like the merging
[15:47:19] <Avenger> i don't have any oggs arund here
[15:47:48] <Avenger> i take a peek at the code, maybe i can spot it by looking
[15:47:57] <fuzzie> the merging code doesn't touch the actionsTable
[15:48:23] <fuzzie> and it works for me in all games, so it seems unlikely, but maybe
[15:49:36] <Avenger> what is the actions table size
[15:49:50] <Avenger> because if it is 400, then it is trivial why it crashes
[15:49:56] <Avenger> i put the new actions at 400
[15:50:12] <Avenger> #define MAX_ACTIONS 400
[15:50:14] <Avenger> LOL
[15:50:27] <Avenger> typical
[15:50:40] <Gekz> why is there a maX
[15:50:42] <fuzzie> i guess the code should probably check that :p
[15:50:48] <Avenger> yes it should
[15:51:09] <Avenger> gekz: there is a max, so it is a simple indexing, we don't want to slow this down
[15:51:15] <Gekz> ah.
[15:51:19] <Gekz> cant it be set at runtime
[15:51:35] <fuzzie> yes, we could use a std::vector, but there is not much point in adding the complication
[15:51:48] <Gekz> ok
[15:51:58] <Gekz> going to sleep now that I've done a subpar audit of your work
[15:51:59] <Gekz> xD
[15:52:00] <Gekz> night
[15:52:01] <Avenger> well, the size could be configurable, especially now
[15:52:09] <fuzzie> i tested with slightly more sensible action ids, i guess :)
[15:52:27] <Avenger> yes, i put it to 400 so it is working with all engines
[15:52:36] <Avenger> but we could count back from 400 :)
[15:52:51] <fuzzie> you want me to add some sanity checks?
[15:53:03] <Avenger> i think i will just count back from 400, yes, add them
[15:53:09] <Avenger> i will just mod gemact then
[15:54:11] <Avenger> ok lets see if the forge opens now :)
[15:57:36] <lynxlynxlynx> Movable Orientation goes the same as the Orient in animations (0-15 starting from south and incrementing clockwise)?
[15:57:49] --> tombhadAC has joined #gemrb
[15:57:55] <fuzzie> lynxlynxlynx: i think so :)
[16:01:22] <CIA-22> gemrb: 03fuzzie * r7339 10/gemrb/trunk/gemrb/plugins/Core/GameScript.cpp: add some sanity checks to trigger/action/object reading
[16:01:45] <fuzzie> now your gemact.ids simply is complained about and ignored :p
[16:02:39] <lynxlynxlynx> bash has a nice feature, you can break/continue outer loops from functions
[16:02:53] <fuzzie> some might argue over 'nice' :)
[16:03:44] <lynxlynxlynx> :P
[16:03:56] <fuzzie> okay, does anything actually use NearSavedLocation?
[16:04:05] <fuzzie> i want to sabotage gemrb's version to always return 1
[16:04:37] <lynxlynxlynx> how and iwd2
[16:04:40] <lynxlynxlynx> and pst
[16:04:41] <fuzzie> hum, drat.
[16:04:44] <fuzzie> maybe i need a pst-specific flag
[16:04:46] <lynxlynxlynx> plenty of matches
[16:05:02] <fuzzie> in pst, it means "are you near the last location that got saved to the CRE struct?", i think
[16:05:24] <lynxlynxlynx> most common argument is 0, but it goes up to 156
[16:05:24] <fuzzie> so gemrb's implementation breaks pst completely
[16:05:36] <fuzzie> but if it's used in iwd/iwd2 then maybe it works there
[16:05:45] <lynxlynxlynx> only how
[16:05:54] <lynxlynxlynx> oh
[16:05:59] <lynxlynxlynx> i don't have an iwd script dump
[16:06:27] <lynxlynxlynx> so put more simply, only bgs don't use it
[16:07:24] <fuzzie> ok, i wonder how to handle that, some new game flag?
[16:08:02] <fuzzie> any idea if it actually works in how?
[16:08:55] <Avenger> meh, now another crash
[16:09:33] <Avenger> for: overDoor->Highlight = false; and overdoor is not 0
[16:09:38] <lynxlynxlynx> nope
[16:10:19] <lynxlynxlynx> it appears it is used to ReturnToSavedLocation
[16:10:38] <fuzzie> ReturnToSavedLocation doesn't actually do anything in pst
[16:10:49] <fuzzie> well, it sometimes seems to move a few pixels but i'm not convinced
[16:11:16] <fuzzie> i suspect it's to do with this CRE location thing, somehow :/
[16:11:58] <Avenger> wow, it works
[16:12:16] <fuzzie> Avenger: you don't delete the door? :) it should be wiped in ChangeMap for the only time doors get deleted
[16:12:53] <Avenger> i moved through an area
[16:13:00] <Avenger> and a door was highlighted
[16:13:43] <fuzzie> and you left the area while it was highlighted? or it just crashed randomly?
[16:14:01] <raevol> fuzzie: i can confirm that scar works now, at least just poking at him, not testing him extensively
[16:14:06] <fuzzie> raevol: cool :)
[16:14:32] <Avenger> i think it was highlighted when i used a console command to move to the forge
[16:15:02] <fuzzie> ah
[16:15:14] <fuzzie> guiscript MoveToArea doesn't do it the proper way, i guess
[16:16:01] <Avenger> ok, now i just have to find out, what area to change and such
[16:17:09] <Avenger> so, if ar6004::forge_on is 1, i have to change... what
[16:17:22] <fuzzie> you have to change four infopoints, three in ar6005 and one in 6011
[16:17:39] <fuzzie> change them from ar6004 to ar6013
[16:17:43] <Avenger> one in 6011 too?
[16:17:53] <Avenger> omg i see
[16:18:18] <fuzzie> and Forge_On is a GLOBAL
[16:18:30] <Avenger> hehe, ok
[16:18:56] <raevol> hmm something you guys fixed makes my characters pwn as level 10s should
[16:19:17] <lynxlynxlynx> something that helps is the strength bonus to damage
[16:19:24] <raevol> that could do it
[16:19:25] <Avenger> hehe
[16:19:52] <raevol> i can't wait for the con bonus to hp regen
[16:23:56] <Avenger> hmm, one slight problem i got, there is no way to find out if i already made these changes, so it will always run
[16:25:16] <Avenger> maybe i will need to add another trigger :)
[16:25:37] <Avenger> or rather, an object
[16:25:57] <Avenger> or dunno, objects cannot have a parameter
[16:26:08] <Avenger> i need to find out if there is an exit to ar6004
[16:26:23] <Avenger> or at least, if the current area is 6005/60011
[16:27:22] <Avenger> but there is no way to find these out, and i may not change the forge_on variable, i guess. i don't want to add new variables either
[16:30:56] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
[16:35:17] <fuzzie> well, i could add a trigger, but that is just as slow
[16:36:10] <fuzzie> but Avenger can check the CURRENT_AREA global!
[16:36:18] <fuzzie> and implement it, for good measure
[16:37:00] <fuzzie> i guess AreaCheck would also work fine, or CurrentAreaIs, etc..
[16:40:18] <CIA-22> gemrb: 03avenger_teambg * r7340 10/gemrb/trunk/gemrb/override/ (5 files in 3 dirs):
[16:40:18] <CIA-22> gemrb: changed gemact to count back from 399 (our engine's current action count limit)
[16:40:18] <CIA-22> gemrb: hack for the forge in IWD/HoW
[16:40:58] --> Avenger has joined #gemrb
[16:41:10] --- ChanServ gives channel operator status to Avenger
[16:41:25] <Avenger> ok, the forge should work now, even if you had an old game
[16:41:29] <fuzzie> :)
[16:41:54] <fuzzie> most iwd gemrb saves are broken, because we dropped the killvar flags :(
[16:42:12] <fuzzie> sometime i should change the code so we make sure to read/write everything, i guess
[16:42:42] <Avenger> there are killvar flags?
[16:42:53] <Avenger> in the area actor blocks?
[16:43:00] <fuzzie> no, in the CRE data
[16:43:22] <Avenger> and we just skipped them?
[16:43:25] <fuzzie> you knew about them, you wrote in the comments that you skipped them :)
[16:43:41] <Avenger> hehe
[16:43:45] <fuzzie> i should write on the IESDP forum about KILL_<race name>_CNT, too
[16:43:54] <Avenger> i know i fixed it for pst
[16:44:37] <fuzzie> yes, all works fine for everything now, i think
[16:44:52] <Avenger> which commit fixed it?
[16:45:19] <Avenger> or which line is it in creimp?
[16:45:43] <fuzzie> search for 'SetDeathVar' and 'IncKillCount'
[16:45:52] <Avenger> i see them
[16:45:57] <fuzzie> i implemented them safely for now, so we don't lose more data
[16:46:27] <fuzzie> but hopefully they are just flags
[16:46:42] <Avenger> well, you use new fields for these
[16:46:55] <fuzzie> then r7322 fixes the Die() code and r7323 fixes the iwd2 DeathVarFormat
[16:47:00] <Avenger> couldn't they be shoehorned into the DeathCounters ?
[16:47:14] <fuzzie> well, i don't know if they're flags.
[16:47:21] <Avenger> pst has 4 fields for deathcounters
[16:47:37] <fuzzie> if someone REs iwd and iwd2 and says they are, then we can set something else
[16:47:53] <fuzzie> but i don't really want to just overwrite pst stuff, it means no-one could use both in mods
[16:48:16] <Avenger> ah yes, pst has numbers there
[16:48:21] <Avenger> that's what you mean?
[16:48:31] <fuzzie> yes, maybe iwd/iwd2 can have numbers there too?
[16:48:39] <fuzzie> anyway, it works now, i leave it alone for now :)
[16:49:02] <fuzzie> oh, and of course the DeathVarFormat change breaks all gemrb iwd/iwd2 savegames too
[16:49:08] <Avenger> ok, but if it is numbers that can actually decrease/increase the variable, then we merge it into the pst style code
[16:50:12] <Avenger> at least it is closer to the truth now
[16:50:15] <fuzzie> dammit, i see a bug in that code already
[16:51:33] <CIA-22> gemrb: 03fuzzie * r7341 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: death variable bugfix
[16:52:09] <fuzzie> i didn't test the race counter code, someone probably should
[16:52:36] <CIA-22> gemrb: 03avenger_teambg * r7342 10/gemrb/trunk/gemrb/plugins/FXOpcodes/FXOpc.cpp: backstab modifier: an odd line which is in the original engine
[16:53:05] <fuzzie> i guess i should try and fix that iwd2 spellbook crash, so i can test it in iwd2
[16:53:25] <Avenger> well, if you wish :)
[16:53:40] <Avenger> also lynx can continue his heroic walkthrough in iwd :)
[16:53:55] <Avenger> kill terikan and save the forge :)
[16:54:06] <fuzzie> if you fixed the forge, iwd works now up until the end
[16:54:16] <fuzzie> the cutscene for the end credits doesn't run, i'm not sure why
[16:54:20] <Avenger> oh it was already done?
[16:54:24] <Avenger> cool
[16:54:42] <Avenger> the complete crystal tower runs?
[16:55:28] <Avenger> it was full of odd stuff, like the pomabs and the belhifet portal
[16:57:22] <fuzzie> ok, i got to eat pizza :)
[16:57:44] <Avenger> see you later !
[16:59:32] <fuzzie> or not, had to put it back in oven for a bit :(
[16:59:52] <Avenger> hehe, buggy pizza
[16:59:55] <fuzzie> and, yes, with all the action fixes i think it's good? but you'd have to ask lynx :)
[17:00:13] <fuzzie> SoA gets better while we fix things, too
[17:00:17] <Avenger> do you have soa saves?
[17:00:22] <Avenger> i think one bug is fixed no
[17:00:24] <Avenger> now
[17:00:34] <Avenger> there's a shielded unnamed thing (rat?) ...
[17:00:43] --> pupnik_ has joined #GemRb
[17:00:46] <Avenger> this is surely a resisted invisibility
[17:00:50] <Avenger> which is now fixed
[17:00:54] <fuzzie> i don't think i have a save around there
[17:01:12] <Avenger> that's the place where you enter the asylum
[17:01:21] <Avenger> or rather, brynnlaw
[17:01:36] <Avenger> well, my soa saves are all before brynnlaw ;)
[17:01:44] <Avenger> maybe i should continue playing
[17:02:27] <fuzzie> i have an 'entering spellhold' save from lynx, but i assume it's too far
[17:02:44] <Avenger> it is too far, yep
[17:03:11] <Avenger> well, i guess all save would work, if you give yourself 20000gc
[17:03:15] <fuzzie> yikes, i can use a travel trigger at far too much a range, now
[17:03:19] <Avenger> i will try that way
[17:03:38] <Avenger> or you have to kill the vampires first?
[17:03:40] <Avenger> hmm
[17:04:09] <fuzzie> you can kill the vampires pretty quick with teleport, heh
[17:09:13] --> barra_away has joined #gemrb
[17:12:38] <Avenger> hey barra, what's up with your game?
[17:20:11] <-- Avenger has left IRC ("bye!")
[17:20:48] <lynxlynxlynx> the pomabs were wierd indeed
[17:21:01] <lynxlynxlynx> i'm not sure if how broken that part is
[17:26:04] <-- barra_library has left IRC (Read error: 110 (Connection timed out))
[17:33:01] <-- pupnik_ has left IRC ("leaving")
[17:36:21] --> pupnik has joined #GemRb
[17:55:54] <fuzzie> lynxlynxlynx: you seem to have accidentally the whole pst
[17:55:59] <fuzzie> Traceback (most recent call last):
[17:55:59] <fuzzie> File "./GUIScripts/pst/GUIWORLD.py", line 308, in OpenFormationWindow
[17:55:59] <fuzzie> GemRB.LoadWindowPack (GetWindowPack())
[17:55:59] <fuzzie> NameError: global name 'GetWindowPack' is not defined
[17:56:36] <lynxlynxlynx> doesn't it import guicommonwindows?
[17:56:54] <lynxlynxlynx> maybe it is too late, try adding an import of that function from guicommon
[17:57:04] <fuzzie> no, it imports specific things from there
[17:57:12] <fuzzie> whoever wrote this was a bit tidier i guess
[17:57:30] <fuzzie> i'll test it with the import
[17:58:05] <fuzzie> that works fine
[17:59:27] <fuzzie> i find pst using stuff like !NearLocation(Myself, [1839.1515], 0) which fails
[17:59:48] <fuzzie> i wonder how to deal with that
[18:00:01] <fuzzie> snapping to a search square, maybe
[18:03:30] <CIA-22> gemrb: 03fuzzie * r7343 10/gemrb/trunk/gemrb/GUIScripts/pst/ (GUIWORLD.py QuitGame.py): add missing GetWindowPack imports to pst guiscript
[18:11:25] <lynxlynxlynx> silly me
[18:11:47] <fuzzie> i think no-one really tests pst :)
[18:11:58] <fuzzie> i am currently being annoyed by areas where everyone is just sitting there, doing nothing
[18:12:08] <lynxlynxlynx> i wasn't specificly complaining about that ;)
[18:13:22] <fuzzie> but this is a combination of bugs
[18:13:30] <fuzzie> i will have to glare at Avenger some more
[18:13:35] <CIA-22> gemrb: 03lynxlupodian * r7344 10/gemrb/trunk/gemrb/includes/strrefs.h: string order is very important
[18:13:50] <fuzzie> because this stance ids thing is really kinda bad :(
[18:16:37] <fuzzie> heh, the pst script calls 'StickySinisterPoof'
[18:19:14] <fuzzie> which also breaks, i guess
[18:19:24] <fuzzie> but the broken animations are the worst problem, i guess
[18:25:38] <raevol> hello hello
[18:27:47] <fuzzie> hi, raevol!
[18:27:52] <raevol> :D
[18:28:09] <fuzzie> very cool that you managed to get so far :)
[18:28:17] <raevol> yes, it's fun
[18:28:35] <raevol> i'm curious about these dopplegangers in the candlekeep tombs though
[18:28:56] <raevol> i have characters from the training coming up to me and saying creepy things, and then turning into dopplegangers
[18:29:00] <raevol> and then they just disapear
[18:29:06] <raevol> i dunno if that's the expected behavior
[18:29:16] <fuzzie> do you have a save somewhere near that?
[18:29:27] <fuzzie> i haven't played bg1 in too long, but i can check the scripting
[18:29:29] <raevol> my quick save that i sent you before is right before the area
[18:29:30] <CIA-22> gemrb: 03fuzzie * r7345 10/gemrb/trunk/gemrb/plugins/Core/Triggers.cpp: sabotage NearSavedLocation for pst
[18:29:45] <fuzzie> oke
[18:29:51] <raevol> you just have to travel northeast to the next area
[18:31:43] <fuzzie> that applies a DOPPLEGANGER_CHANGE_DEFAULT_LESSER spell, turns into dopple.cre, but indeed instant death, odd
[18:32:07] <raevol> hmm
[18:32:35] <fuzzie> i'll check the obvious a moment
[18:33:31] <fuzzie> heh
[18:33:39] <fuzzie> so Avenger added an unsummon to the replace creature effect :-)
[18:33:57] <raevol> so this is much less an issue, but on that same map there's elminster, and he's a doppleganger, but after he turns into a doppleganger the elminster character stays there and is just silent
[18:34:01] <raevol> haha nice
[18:35:08] <fuzzie> waiting for svn..
[18:35:25] <fuzzie> maybe we should host it somewhere else :/
[18:35:30] <raevol> haha
[18:37:46] <CIA-22> gemrb: 03fuzzie * r7346 10/gemrb/trunk/gemrb/plugins/FXOpcodes/FXOpc.cpp: don't unsummon replacement creatures
[18:37:47] <fuzzie> ok, try that, and see if elminster is still broken? if so, i'll look
[18:37:51] <raevol> ok
[18:45:00] <raevol> the doppleganger vanishing is fixed, but elminster remains
[18:46:46] <fuzzie> ok :)
[18:48:08] <CIA-22> gemrb: 03lynxlupodian * r7347 10/gemrb/trunk/gemrb/ (6 files in 4 dirs):
[18:48:08] <CIA-22> gemrb: added "proper" backstab support - in tob you have to actually target the
[18:48:08] <CIA-22> gemrb: back for a valid /back/stab attempt
[18:51:16] <raevol> back in a bit
[18:52:30] <fuzzie> huh, i get an action id of 4192 from elminster's dialog
[18:53:32] <fuzzie> hmm
[18:53:39] <lynxlynxlynx> bitfield?
[18:53:55] <fuzzie> yes, i guess it's an ActionOverride
[18:54:11] <fuzzie> and the override is handled in the wrong place
[18:55:28] <fuzzie> our GenerateAction just doesn't work for it, i guess?
[18:55:36] <lynxlynxlynx> don't look at me
[18:55:43] <fuzzie> :_)
[18:58:01] <lynxlynxlynx> re the brynlaw rat: it is gone
[18:58:08] <lynxlynxlynx> but the protection globe remains
[18:58:26] <fuzzie> heh, i guess we should also draw VVCs with that trans setting
[18:58:55] <fuzzie> oh, we should .. DrawVideocells gets the tint
[19:00:30] <fuzzie> i guess the problem is, the VVC Draw routine only pays attention to the tint if it has IE_VVC_TINT set
[19:00:43] <fuzzie> try disabling the DrawVideocells call if the actor is invis?
[19:01:23] --> Maighstir has joined #gemrb
[19:03:23] <lynxlynxlynx> well, completely invisible, not just translucent, then it may be ok
[19:03:36] <fuzzie> yes, i mean, check the state
[19:04:40] <lynxlynxlynx> and I can also check for the dead one, which would fix another bug :)
[19:05:59] <fuzzie> i think Avenger said that's wrong
[19:06:06] <fuzzie> that some VVCs still render after death
[19:06:44] <fuzzie> but maybe i misremember, don't know
[19:07:38] <lynxlynxlynx> i'll ask him when he's back
[19:14:00] <lynxlynxlynx> yeah that worked
[19:16:42] <lynxlynxlynx> forge works too :)
[19:17:05] <fuzzie> oops, Elminster crash is my bad
[19:18:36] <fuzzie> i hadn't expected scripts to ActionOverride themselves!
[19:18:39] <CIA-22> gemrb: 03fuzzie * r7348 10/gemrb/trunk/gemrb/plugins/Core/GameScript.cpp: don't release an action we still need
[19:19:20] <-- Maighstir has left #gemrb ()
[19:21:01] --> Maighstir has joined #gemrb
[19:22:31] <lynxlynxlynx> and i didn't kill the barbarian threat
[19:22:34] <fuzzie> hm, Imoen seems permanently Held in this spider web
[19:23:02] <lynxlynxlynx> made also fighters immune to backstab ><
[19:23:17] <fuzzie> knwon bug?
[19:23:34] <lynxlynxlynx> web? no
[19:25:18] <fuzzie> spwi215.spl from bg1, loads a gemrb web.pro
[19:32:06] <fuzzie> since gemrb ignores my ctrl-m, i'm not sure how to debug
[19:32:47] <lynxlynxlynx> check which duration it has set
[19:32:57] <fuzzie> type 0 [duration], duration 7
[19:34:49] <lynxlynxlynx> odd
[19:35:38] <fuzzie> i don't understand enough of effects to see where it's meant to get undone
[19:35:59] <fuzzie> oh, i guess SetSpellState twiddles Modified
[19:38:18] <fuzzie> it's reproducible every time, at least
[19:39:27] <fuzzie> but i don't see any way of seeing what's going on
[19:41:23] <lynxlynxlynx> add a target->DebugDump() call somewhere on the path?
[19:41:39] <fuzzie> but where's the path, i wonder
[19:42:14] <lynxlynxlynx> Payload perhaps
[19:45:17] <fuzzie> i just hacked ctrl-m to give me the damn actor
[19:47:54] <fuzzie> 10: 0x9d: Overlay:Web (0, 0) S:spwi215
[19:47:54] <fuzzie> 11: 0x6d: State:Hold3 (0, 2) S:spwi215
[19:48:11] <fuzzie> make any sense to you?
[19:49:41] <lynxlynxlynx> other hold states do exhibit complete blackouts
[19:49:58] <lynxlynxlynx> but yours is permanent
[19:50:16] <lynxlynxlynx> check the bg2 web if it uses the same state
[19:50:28] <fuzzie> there are two different FXOpc codes for this
[19:50:39] <fuzzie> one marked 'iwd2' which this is using
[19:51:00] <fuzzie> but the normal one for this opcode seems to be fx_hold_creature_no_icon
[19:52:55] <fuzzie> so i guess gemrb's effects.ids is bad?
[19:52:58] <fuzzie> i never knew it even existed!
[19:55:16] <fuzzie> bg2's web seems to be identical
[19:55:37] <fuzzie> except it doesn't use gemrb's web.pro
[19:57:07] <CIA-22> gemrb: 03fuzzie * r7349 10/gemrb/trunk/gemrb/plugins/Core/GameControl.cpp: make ctrl-m poke through the actor list manually to catch immobile targets
[19:57:08] <fuzzie> i'
[19:57:13] <fuzzie> i'm sure there's a better fix, but i'm tired
[20:03:32] <fuzzie> they're timing mode 10, duration already before gametime
[20:04:12] <fuzzie> so that's FX_DURATION_JUST_EXPIRED, which .. makes sense
[20:04:50] <fuzzie> but Avenger put the fxqueue Cleanup in UpdateActorState
[20:04:57] <fuzzie> which isn't called on held characters, because they're held.
[20:05:50] <fuzzie> someone's added that since i last touched this code, actually.
[20:06:02] <fuzzie> actor->Immobile -> don't update effects -> don't make actor mobile? genius :)
[20:07:15] <lynxlynxlynx> :)
[20:10:11] <fuzzie> all fixed now
[20:11:05] <CIA-22> gemrb: 03fuzzie * r7350 10/gemrb/trunk/gemrb/plugins/Core/ (Actor.cpp Map.cpp): play musical chairs with the fxqueue.Cleanup call
[20:16:19] <fuzzie> [WorldMap]: The worldmap file is corrupted... and it would crash right now!
[20:16:21] <fuzzie> Segmentation fault
[20:16:25] <fuzzie> ^- Oh.
[20:29:31] <CIA-22> gemrb: 03fuzzie * r7351 10/gemrb/trunk/gemrb/plugins/Core/WorldMap.cpp: avoid a segfault when we can't find a worldmap icon
[20:38:23] <CIA-22> gemrb: 03fuzzie * r7352 10/gemrb/trunk/gemrb/plugins/Core/GSUtils.cpp: avoid segfault on missing cre file
[20:57:24] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[21:04:55] <-- tombhadAC has left IRC ("Verlassend")
[21:30:29] --> Avenger has joined #gemrb
[21:30:33] --- ChanServ gives channel operator status to Avenger
[21:31:24] <Avenger> fuzzie: i see you moved Cleanup, i hoped it will fix the equipping refresh bug, but it didn't.
[21:32:27] <fuzzie> when i looked, it seemed that the python just didn't pay attention to the flag enough
[21:35:50] <Avenger> no, the stats are not refreshed, i already seen that
[21:36:00] <Avenger> i don't know why, it is nonsense :)
[21:36:53] <fuzzie> huh :(
[21:38:07] <CIA-22> gemrb: 03fuzzie * r7353 10/gemrb/trunk/gemrb/plugins/CREImporter/CREImp.cpp: seek to correct offsets when loading iwd2 creatures
[21:38:14] <fuzzie> that took me an hour to spot the obvious :(
[21:40:31] <Avenger> but you did :) hehe, what did this fix?
[21:41:12] <Avenger> the spellbook crasher?
[21:41:14] <fuzzie> yes
[21:41:21] <Avenger> nice!
[21:41:43] <fuzzie> some script breaks once i start playing, so i look at that now
[21:41:55] <Avenger> that means a big advance in iwd2 compatibility
[21:45:50] <fuzzie> ah, this script sets IE_DIALOGRANGE and we don't use that, i guess
[21:46:51] <Avenger> that thing isn't completely understood
[21:47:49] <Avenger> i don't quite see why initiating dialogs would need a stat, you can always use a range trigger in the dialog starter script.
[21:47:59] <Avenger> But i have no other idea what it might do
[21:48:30] <fuzzie> well, this script seems to be using it as a way to make sure the dialog starts
[21:48:42] <fuzzie> so that the actor doesn't try walking to the target because it's far away
[21:49:25] <fuzzie> don't see what's wrong with using an action that doesn't check distance, though :)
[21:50:33] <fuzzie> i guess iwd2 doesn't have very many dialog actions, but StartDialogueNoSet seems like it would do. maybe it checks distance too in iwd2.
[21:55:04] <Avenger> ahh, so it is using it to set the minimum distance, well that is easy to implement
[21:55:26] <Avenger> simply set the dialog distance in the creature at load time, and use the stat instead of the constant
[21:55:33] <fuzzie> makes sense :)
[22:00:53] --> tombhadAC has joined #gemrb
[22:50:17] <-- Avenger has left IRC ("bye!")